Someone selling you a copper peptide is selling you a photograph. The published GHK-Cu results come from two human trials, one from 1992 and one from 2006, and neither produced a significant objective improvement over a placebo cream or a standard one. The people searching this phrase deserve the two papers instead of the slide deck, so here is what the papers actually report.
The direct answer: 13 people were treated after laser skin resurfacing and 86 people were treated for venous leg ulcers, and in both studies the objective measures tied or lost to the comparison group. The one measure that favored GHK-Cu was a patient satisfaction questionnaire in the 13-person study, significant at P = 0.04.
We read the indexed abstracts and the trial designs they describe. We did not read the full text behind the journals’ paywalls, and we say so rather than dress an abstract up as a manuscript.
GHK-Cu results in the human record: two trials, 13 people and 86 people
A PubMed search for GHK-Cu, also indexed as glycyl-histidyl-lysine and as the copper tripeptide complex, filtered to the clinical trial and randomized controlled trial publication types, returns three records. One of the three is a study in rabbits. The other two are the studies below, from the Journal of Vascular Surgery in August 1992 and the Archives of Facial Plastic Surgery in July-August 2006.
That count is the honest shape of the evidence, and it needs one caveat. Publication-type indexing is not a guarantee that no other human trial exists. A study can sit unindexed, or run and never be published. What the count does show is that a clinic quoting “clinical studies” is quoting from a set this small.
The 2006 skin trial: the GHK-Cu results moved on a questionnaire, not on the photographs
Miller and colleagues, in the Archives of Facial Plastic Surgery of July-August 2006, took patients having circumoral skin resurfacing with a carbon dioxide laser at standard settings and randomized them to a post-treatment skin regimen with or without the copper tripeptide complex.
Thirteen patients completed the study. Erythema was scored across the post-treatment period by computer software and by blinded evaluators. Twelve weeks after treatment, overall improvement in wrinkles and overall improvement in skin appearance were assessed, and patients filled in a validated questionnaire before treatment and twelve weeks after.
The computer analysis and the blinded evaluators found no statistically significant difference between the groups for earlier resolution of erythema. All the patients, in both groups, had significant improvement in wrinkles and overall skin quality, and there was no difference between the groups. Then the questionnaire came back and reported a significant difference in the post-treatment improvement of overall skin quality for the patients using GHK-Cu, at P = 0.04.
Read that twice, because it is the whole argument in miniature. The instrument that could not see the treatment was the blinded one. The instrument that saw it was the one that knew which cream the patient had been given. The abstract reports the P value and stops. It does not report any test of why satisfaction moved.
The 1992 ulcer trial: the copper cream tied the placebo and lost to a silver cream
Bishop and colleagues, in the Journal of Vascular Surgery of August 1992, ran a prospective randomized evaluator-blinded trial in patients with venous stasis ulcers, a condition whose recurrence rate the paper puts at close to 70 percent even after successful initial management. Eighty-six evaluable patients completed the trial, comparing silver sulfadiazine 1 percent in a cream, a biologically active tripeptide copper complex 0.4 percent cream, and an inert vehicle placebo.
Silver sulfadiazine statistically reduced ulcer size compared with both the copper complex cream and the placebo. There was no difference between the copper cream and the placebo.
The dose and the route matter for anyone reading this against an injectable sold online. This was a 0.4 percent cream applied to the wound, and it still did not separate from a cream containing nothing active.
The third trial in the index used 18 rabbits, and it is the one that won
Cangul and colleagues, in Veterinary Dermatology of December 2006, made full thickness wounds on both sides of the dorsal midline in 18 New Zealand rabbits, then applied topical tripeptide-copper complex, zinc oxide, or nothing, daily for 21 days. The unhealed wound area was significantly smaller in the copper group than in the untreated control group on days 7, 14 and 21, and filling of the wound with granulation tissue was significantly slower in the control group.
This is the study that supports the phrase wound healing, and it is a rabbit study. It is the only member of the set that reported a clear win over an untreated control, and it was not run in humans.
FDA’s position: under evaluation, and the injectable route carries the warning
FDA’s document on bulk drug substances nominated for compounding under section 503A, updated May 14, 2026, lists “GHK-Cu (except for injectable routes of administration)” in Category 1, the agency’s category for bulk drug substances still under evaluation. The document explains the sequence: GHK-Cu was removed from category 1 on April 22, 2026 because the nominations were withdrawn, and on May 5, 2026 one nominator clarified that it had intended to withdraw only the injectable route, so the substance was added back to category 1 for non-injectable routes. FDA states it intends to consult the Pharmacy Compounding Advisory Committee before the end of February 2027 about whether GHK-Cu belongs on the 503A bulks list at all.
Category 1 is not approval and it is not inclusion. It is a queue.
The injectable route sits in a different document. FDA’s page on bulk drug substances that may present significant safety risks, current as of April 22, 2026, carries GHK-Cu for injectable routes of administration in its table of substances nominated and then withdrawn, with this assessment: “Compounded injectable drugs containing GHK-Cu may pose risk for immunogenicity due to the potential for aggregation and peptide-related impurities. There are limited data in humans to inform safety-related considerations.”
Two things follow. FDA wrote the immunogenicity concern about the injected form, and the agency’s own words put the human data in the limited bucket, which is the same thing these two trials say from the other direction.
What a before-and-after photograph cannot tell you
A photograph has no control group, no blinding, no sample size and no time point. Those four things are the difference between a result and an anecdote, and the two human trials above are what happens when you insist on them.
There is a fair question hidden in the 2006 questionnaire, and it cuts at the seller rather than the buyer. If satisfaction rises while blinded scoring does not move, one candidate explanation is that the person applying an expensive cream expects it to work. The study does not test that. It reports the numbers, and the numbers say the visible change was not there.
Anyone shopping for a clinic should be able to ask one question and get a straight answer: which trial, how many people, who scored the outcome, and what did the comparison group get. A clinic that answers with a paper is worth a conversation. A clinic that answers with a reel is answering a different question.
A prescription decision belongs to a prescriber with your history and your labs. Our clinic directory lists practices by city, and we do not check licenses, so verify each clinic’s license with the state medical board before you book. Our compound pages cover the peptides that come up most often in clinic visits, including the FDA status we can document for each.
Frequently asked questions
Did GHK-Cu work in human trials?
Not on any objective measure reported. In the 2006 study of 13 patients after laser resurfacing, blinded evaluators and computer analysis found no significant difference in erythema resolution and no significant difference in wrinkles or skin quality. In the 1992 study of 86 patients with venous stasis ulcers, the copper complex cream did not differ from the placebo cream, and silver sulfadiazine 1 percent outperformed both. One questionnaire in the 2006 study favored GHK-Cu at P = 0.04.
How many human trials of GHK-Cu are there?
A PubMed search filtered to clinical trial and randomized controlled trial publication types returns three records, and one of them is a study in 18 rabbits. That leaves two human trials, from 1992 and 2006. Publication-type indexing is an imperfect count and cannot prove a negative, which is why we state the search rather than a definitive number.
Is GHK-Cu FDA approved?
No. FDA’s 503A document updated May 14, 2026 places GHK-Cu, excluding injectable routes, in Category 1, which is the category for bulk drug substances under evaluation. FDA says it intends to consult the Pharmacy Compounding Advisory Committee before the end of February 2027 about whether the substance should be on the 503A bulks list. Being under evaluation is not approval.
What were the GHK-Cu results for skin?
In the 2006 Archives of Facial Plastic Surgery study, all 13 patients improved in wrinkles and overall skin quality, and there was no difference between the GHK-Cu group and the group without it. The difference appeared only in the patient questionnaire, at P = 0.04, on overall skin quality after treatment.
Does GHK-Cu help wound healing?
The study usually cited for that claim is the 2006 Veterinary Dermatology study of 18 New Zealand rabbits, where the copper complex beat the untreated control on wound area and granulation tissue. The human wound study is the 1992 venous stasis ulcer trial, and there the copper cream tied the placebo and lost to silver sulfadiazine 1 percent.
Is injectable GHK-Cu legal to compound?
FDA’s page on bulk drug substances that may present significant safety risks, current as of April 22, 2026, lists GHK-Cu for injectable routes of administration among substances nominated and then withdrawn, and states that compounded injectable drugs containing GHK-Cu may pose risk for immunogenicity due to aggregation and peptide-related impurities. The agency also states there are limited data in humans to inform safety considerations. Whether a particular pharmacy may compound a given substance is a question for that pharmacy and the state board that licenses it.
Where can I read the trials?
All three are on PubMed and linked below: Miller and colleagues in the Archives of Facial Plastic Surgery, July-August 2006; Bishop and colleagues in the Journal of Vascular Surgery, August 1992; and Cangul and colleagues in Veterinary Dermatology, December 2006. We read the indexed abstracts on November 1, 2026.
Sources:
- Miller TR, Wagner JD, Baack BR, Eisbach KJ, “Effects of topical copper tripeptide complex on CO2 laser-resurfaced skin,” Archives of Facial Plastic Surgery, July-August 2006, PMID 16847171.
- Bishop JB, Phillips LG, Mustoe TA, and colleagues, “A prospective randomized evaluator-blinded trial of two potential wound healing agents for the treatment of venous stasis ulcers,” Journal of Vascular Surgery, August 1992, PMID 1495150.
- Cangul IT, Gul NY, Topal A, Yilmaz R, “Evaluation of the effects of topical tripeptide-copper complex and zinc oxide on open-wound healing in rabbits,” Veterinary Dermatology, December 2006, PMID 17083573.
- FDA, “Bulk Drug Substances Nominated for Use in Compounding Under Section 503A,” updated May 14, 2026.
- FDA, “Certain Bulk Drug Substances for Use in Compounding that May Present Significant Safety Risks,” content current as of April 22, 2026.
- PubMed search, GHK-Cu or glycyl-histidyl-lysine, filtered to clinical trial and randomized controlled trial publication types.
This article is general information for people comparing peptide clinics and treatment claims. It is not medical advice, it is not a recommendation for or against any product, and it carries no dosing guidance. Compounded preparations are not FDA-approved drugs. Talk with a licensed prescriber about your own history before starting, stopping or changing anything.